- Email: [email protected]
Visible light photoactivity enhancement via CuTCPP hybridized g-C3N4 nanocomposite
Applied Catalysis B: Environmental 166–167 (2015) 366–373
Contents lists available at ScienceDirect
Applied Catalysis B: Environmental journal homepage: www.elsevier.com/locate/apcatb
Visible light photoactivity enhancement via CuTCPP hybridized g-C3 N4 nanocomposite Daimei Chen a,∗ , Kewei Wang a,b , Wangzhi Hong a,b , Ruilong Zong b , Wenqing Yao b , Yongfa Zhu b,∗ a Beijing Key Laboratory of Materials Utilization of Nonmetallic Minerals and Solid Wastes, National Laboratory of Mineral Materials, School of Materials Sciences and Technology, China University of Geosciences, Beijing 100083, China b Department of Chemistry, Tsinghua University, Beijing, 100084, PR China
a r t i c l e
i n f o
Article history: Received 5 October 2014 Received in revised form 21 November 2014 Accepted 24 November 2014 Available online 29 November 2014 Keyword: Photocatalysis g-C3 N4 Sensitization Porphyrin
a b s t r a c t Cu (II) meso-Tetra (4-carboxyphenyl) porphyrin (CuTCPP) hybridized g-C3 N4 (CuTCPP/g-C3 N4 ) composites have been facilely synthesized through the ethanol dispersion method. CuTCPP molecules as a sensitizer could easily assemble on the surface of g-C3 N4 nanosheets mainly through – stacking interaction. The CuTCPP/g-C3 N4 composites show much higher photocatalytic activity for phenol degradation than pure g-C3 N4 under visible light irradiation. The optimum photocatatlytic activity of the CuTCPP/g-C3 N4 composites with weight ratio of CuTCPP at 0.75% is almost 2.2 times as high as that of pure g-C3 N4 under the visible light. The enhancement of the visible light photocatalytic activity comes from the efficient electrons transfer from photoexcited CuTCPP molecules to g-C3 N4 sheets and more efficiently visible-light harvesting due to the CuTCPP sensitization. Both the holes and ·O2− are main oxidative species of CuTCPP/g-C3 N4 for phenol degradation under visible light irradiation. Finally, the possible charge transfer mechanism of enhanced visible light photocatalytic activity was proposed. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Graphite-like carbon nitride (g-C3 N4 ), a metal-free polymer semiconductor, has attracted plenty of attention due to its potential application in solar energy conversion, photosynthesis, electrocatalysis, bioimaging and biomedical application [1–4]. Since Wang et al. firstly reported that the polymeric g-C3 N4 shows superior H2 or O2 evolution activities via water splitting under visible light irradiation, it is considered as an especially promising photocatalyst for hydrogen gas production and organic pollutants degradation using solar energy due to its simple fabrication, low-cost and high stability under light irradiation in solution with pH = 0–14 [5,6]. However, the bandgap of g-C3 N4 is still as large as 2.7 eV with an absorption edge just at 450 nm, which largely restricts the visible-light utilization efficiency. In addition, the high degree of recombination between photoinduced electrons and holes also blocks its practical application. To overcome these problems, many methods have been proposed to improve the photocatalytic performance of g-C3 N4 , for example, by doping nonmetallic element
∗ Corresponding authors. Tel.: +86 10 82332274; fax: +86 10 82322974. E-mail addresses: [email protected] (D. Chen), [email protected] (Y. Zhu). http://dx.doi.org/10.1016/j.apcatb.2014.11.050 0926-3373/© 2014 Elsevier B.V. All rights reserved.
[7], and coupling g-C3 N4 with metals [8], organic semiconductors [9,10], graphene [11,12], activated by organic dyes [13]. Among the various strategies for visible-light harvesting, dye sensitization is an efficient and widely used route to extend the spectral response region of wide-band-gap semiconductors [14–16]. Many dye sensitized photocatalysts have been developed for H2 production, such as sensitized semiconductors [17,18], sensitized multiwall carbon nanotubes [19] and sensitized reduced graphene oxide [20]. Obviously, dye sensitization of g-C3 N4 can also be an efficient way to enhance the utilization efficiency of visible light. Until now, several dyes such as Erythrosin B (ErB), Eosin Y (EY), poly(3-hexylthiophene) (P3HT) and zinc phthalocyanines (ZnPc) were used as sensitizers of g-C3 N4 to enhance the photocatalytic H2 production activity with considerable visible light utilization efficiency [13,21–23]. Porphyrins as light harvesting materials play an important role in photosynthesis. As they possess good chromophore activities over the solar spectrum and good electron donating properties due to their large p-electron systems, porphyrin compounds have been widely regarded as excellent photosensitizers for photocatalysts. For example, the metalloporphyrin-modified TiO2 system has been reported for the photocatalytic degradation of various kinds of organic pollutants in water under visible light irradiation [24–26]. In generally, metalloporphyrin modified photocatalysts
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
were prepared by covalent interaction between the different functional groups ( COOH, OH, etc.) in the periphery of the porphyrin ring and the surface of photocatalysts. It is widely accepted that the covalent bands can form channels, which can enhance the electron transfer between porphyrins and photocatalysts, and further lead to the delocalization of p* orbits of porphyrins [25]. However, the amount of porphyrins absorbed on the surface of the photocatlaysts is limited because most of porphyrins are difficult to covalently interact with photocatlaysts. Zhu et al. prepared porphyrin uncovalently functionalized reduced graphene oxide (RGO) nanocomposite by immobilizing porphyrin on RGO nanosheets via – stacking interactions. Porphyrin moiety in the nanocomposite acts as a photosensitizer and harvesting irradiation light. This composite shows remarkable enhanced photocatalytic activity under UV–vis light irradiation due to the fast electrons transfer from photoexcited porphyrin molecules to RGO sheets [27,28]. Considering the g-C3 N4 has analogue structure with RGO (planar layers with electron conjugation), it is expected to interact strongly with porphyrins through electrostatic interaction, – stacking, or hydrogen bonding. However, to the best our knowledge, there are no studies about the sensitization of g-C3 N4 with porphyrin, especially the using the dye sensitized g-C3 N4 for the organic containment degradation. Therefore, a systemic investigation of dye sensitized g-C3 N4 and the oxidation mechanic of the dye sensitized g-C3 N4 for the photodegradation needs further studied. In our paper, we reported the preparation, characterization, and photocatalysis of CuTCPP/g-C3 N4 composites. Choosing CuTCPP as a sensitizer to hybridize with g-C3 N4 is not only due to that porphyrin can easily hybridize with g-C3 N4 through strong – stacking interactions, but also due to its nice locally conjugated aromatic structure, as well as its remarkable chemical and thermal stability benefitting for photocatalytic reactions. The optical absorption of g-C3 N4 (<460 nm) can be greatly extended into longer wavelength of visible light region (450–600 nm) by spectral sensitization with porphyrin. This sensitization photocatalyst shows a high activity for phenol degradation under visible light irradiation because the photogenerated electrons of the excited porphyrin can transfer to g-C3 N4 efficiently. The present results show the promising application of porphyrin derivatives in a photocatalytic organic degradation system for more efficiently utilizing solar radiation. 2. Experimental 2.1. Material Dicyandiamide (C2 H4 N4 ) was purchased from Sinopharm Chemical Reagent Corp, P. R. China. Cu (II) meso-Tetra (4carboxyphenyl) porphine (CuTCPP, C48 H28 CuN4 O8 ) was supported by J&K Scientific Ltd. All other reagents used in this research were analytically pure and used without further purification. 2.2. Sample preparation 2.2.1. Synthesis of g-C3 N4 Polycondensation of dicyandiamide was employed to obtain gC3 N4 according to Ref [29]. Typically, 15 g of dicyandiamide was put in a Muffle Furnace and heated to 550 ◦ C for 4 h to complete the reaction. Finally, the yellow products were collected and ground into powder in an agate mortar. 2.2.2. Synthesis of CuTCPP/g-C3 N4 composite photocatalyst The CuTCPP/g-C3 N4 photocatalysts were prepared by liquid ultrasonic route in ethanol. Firstly, 0.5 g g-C3 N4 was added into ethanol and then was placed in an ultrasonic bath for an hour to completely disperse the g-C3 N4 . The CuTCPP ethanol solution (0.1 mg mL−1 ) was added into the above solution and sonicated for
367
an hour. Then the resulting suspension was continually stirred in a fume hood at 60 ◦ C for a night. After completely volatilization of the ethanol, the samples were dried at 100 ◦ C for 8 h. A series of CuTCPP/g-C3 N4 photocatalysts with the different mass ratios of CuTCPP were prepared by this method, which were denoted as X%CuTCPP/g-C3 N4 , where X represents the weight ratios of CuTCPP in total (X% = 0.1%, 0.25%, 0.5%, 0.75%, 0.8%, 0.9%, 1%, 2%). 2.3. Characterization The crystallinity of the as-prepared sample was characterized by X-ray diffraction (XRD) on Bruker D8 Advance X-ray diffractometer at room temperature. Morphologies and structures of the prepared samples were further examined with a HITACHI HT770 transmission electron microscopy (TEM) operated at an accelerating voltage of 100 kV. The UV–vis diffuse reflectance spectra (DRS) of the samples were carried out on a Hitachi U-3010 UV–vis spectrophotometer using BaSO4 as the reference. Fourier transform infrared (FTIR) spectra were carried on a Bruker spectrometer in the frequency range of 4000–600 cm−1 with a resolution of 1 cm−1 . Electrochemical and photoelectrochemical measurements were performed in three-electrode quartz cells with a 0.1 M Na2 SO4 electrolyte solution. Platinum wire was used as the counter electrode, and saturated calomel electrodes (SCE) were used as the reference electrodes, respectively. The as-prepared photocatalyst film electrodes on ITO served as the working electrode. The photocurrents were measured on an electrochemical system (CHI-660B, China). Electrochemical impedance spectra (EIS) were measured at 0.0 V. A sinusoidal ac perturbation of 5 mV was applied to the electrode over the frequency range of 0.05–105 Hz. Photoluminescence spectra (PL) of the samples were obtained at room temperature excited by incident light of 375 nm using a HORIBA Aqualog Fluorescence Spectrometer. 2.4. Photodegradation experiment The photocatalytic activities of the as-prepared samples were evaluated by the decomposition of phenol under visible light irradiation. Visible light source was obtained by a 500 W Xe lamp with different wavelengths (420 nm, 450 nm, 470 nm, 490 nm, 510 nm, 550 nm) cutoff filter and, and the average light intensity was 35 mW/cm2 . 25 mg photocatalyst was added into prepared 50 ml 5 ppm phenol solution. Before the light irradiation, the suspensions were firstly ultrasonic dispersed in dark for 15 min, then magnetically stirred for 120 min to reach the adsorption–desorption equilibrium. At given time intervals, 2.8 ml aliquots were sampled and filtered with a micropore membrane. Synchronously, the filtrates of phenol solutions at different conditions were analyzed by recording variations of the phenol peak area. HPLC was adopted to detect phenol concentration and its degradation products distribution. Methanol and water (volume ratio: 60/40) were mobile phase, the elution time was 7 min, the flow rate was 1 ml/min, determine wavelength was 270 nm and chromatographic column was a Venusil XBP-C18 (3.9 × 200, Agela Technologies Inc.) column. Before the analysis, the samples were filtered through Millipore discs of 0.45 m to protect the chromatographic column. 3. Results and discussion 3.1. Morphology and Structure of CuTCPP hybridized g-C3 N4 photocatalysts Experimentally, the formation of CuTCPP hybridized g-C3 N4 structure was involved involves a two-steps process (Scheme 1). First, g-C3 N4 is exfoliated into sheet structures through an ultrasonic method in ethanol solvent. Secondly, the CuTCPP ethanol solution (0.1 mg mL−1 ) was added into the above solution and
368
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
Scheme 1. Schematic illustration of the preparation of the CuTCPP hybridized g-C3 N4 composites.
bath-sonicated for an hour. Consequently, the CuTCPP molecules will spontaneously coat on the surface of g-C3 N4 nanosheets through the intermolecular interactions, such as electrostatic interaction and – stacking interaction, resulting in a CuTCPP/g-C3 N4 nanohybrid composite. It is well-known that porphyrin molecules, which have a two-dimensional 18--electron aromatic structure, could interact with g-C3 N4 through – stacking. Meanwhile because the functional groups–COOH in the periphery of the CuTCPP ring might interact with the groups NH2 on the edge of the g-C3 N4 , the CuTCPP and g-C3 N4 molecules can also be covalently bonded together by an amide bond. Therefore, both the uncovalently and covalently interactions facilitate the formation of the CuTCPP/g-C3 N4 hybrid nanocomposite. However, owing to the small amount of the groups NH2 on the edge of the g-C3 N4 , the covalent interaction between the CuTCPP and g-C3 N4 molecules might be very little. So, it can be deduced that the hybridization effect between the CuTCPP and g-C3 N4 molecules might mainly come from the noncovalent interaction. The great advantage of the strategy of noncovalent interaction is that it can combine the unique properties of the CuTCPP and g-C3 N4 , while it does not disturb the physical properties of both moieties [30]. Fig. 1 shows transmission electron microscopy (TEM) images of pure g-C3 N4 , CuTCPP and 0.75%CuTCPP/g-C3 N4 . Most of irregular and large plates of g-C3 N4 are observed in Fig. 1a and b. In the image of CuTCPP (Fig. 1 c and d), a lot of small rodlike structure materials can be observed which might be the CuTCPP molecules aggregations. The average size of CuTCPP molecules aggregations is about 50 × 200 nm. For the 0.75%CuTCPP/g-C3 N4 composite (Fig. 1e and f), there are many large plates of g-C3 N4 , while no CuTCPP aggregation can be found in the TEM image. Meanwhile, compared with pure g-C3 N4 , the agglomeration of CuTCPP/g-C3 N4 composite becomes more serious. These observations demonstrate that CuTCPP molecules might be well dispersed and attached on g-C3 N4 plates. To identify the change of phase structures before and after hybridization, the XRD patterns of the as-prepared CuTCPP/g-C3 N4 composites, as well as pure g-C3 N4 and CuTCPP as shown in Figure S1. The g-C3 N4 sample has a characteristic peak at 27.4◦ , which
can be indexed as the (0 0 2) diffraction plane. And it is can be seen that the crystal phase of g-C3 N4 did not change with increasing CuTCPP content. There was no crystalline CuTCPP in the CuTCPP/gC3 N4 photocatalysts. This fact might be that the loading amount of CuTCPP is too low to detect, and it can also be inferred that CuTCPP molecules dispersed uniformly on the surface of g-C3 N4 , which exhibits a short-range order of CuTCPP molecules arrangement. The FT-IR spectra of pure g-C3 N4 , pure CuTCPP and a series of CuTCPP/g-C3 N4 photocatalysts are shown in Fig. 2. In the FT-IR spectrum of g-C3 N4 , the peak at 1637 cm−1 is attributable to the C N stretching vibration modes, while the peaks at 1243 cm−1 , 1320 cm−1 and 1403 cm−1 are due to the aromatic C N stretching [31–33]. The peak at 808 cm−1 is related to the s-triazine ring modes [32]. These peaks are present in the all CuTCPP/g-C3 N4 composites, suggesting that no structural change of g-C3 N4 occurs. It can be seen that there was no characteristic peaks of CuTCPP in the CuTCPP/gC3 N4 photocatalysts at the low CuTCPP loadings (<2%). When the loading amount of CuTCPP is more than 10%, the main characteristic peaks of CuTCPP molecules can be observed clearly, indicating that the CuTCPP molecules might homogenously immobilized on the surface of the g-C3 N4 nanosheets and successfully hybridized with g-C3 N4 . However, the characteristic stretching and stretching band of the amide group formed between the CuTCPP and g-C3 N4 molecules, which are reported to occur at 1640 and 1260, respectively, can not be found in the FT-IR spectra of CuTCPP/g-C3 N4 composite. The likely reason might be that these peaks are too small to be overlapped by the peaks of g-C3 N4 . This further confirms that the hybridization effect between the CuTCPP and g-C3 N4 molecules might mainly come from the noncovalent interaction. 3.2. Photocatalytic activity and photocurrent of CuTCPP hybridized g-C3 N4 composite photocatalysts The photocatalytic performance was mainly evaluated by degradation of phenol, a hazardous pollutant for water as well as a common model to test the photodegradation activity. As shown in Fig. 3a, significant differences in the catalytic behaviors were
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
369
Fig. 1. TEM images of (a, b) pure g-C3 N4 , (c, d) pure CuTCPP, (e, f) 0.75%CuTCPP/g-C3 N4 .
observed, and the photodegradation process was fit to pseudofirst-order kinetics in which the value of rate constant k is equal to the corresponding slope of the fitting line. The blank test confirms that phenol is only slightly degraded in the absence of catalysis, indicating that the photolysis effects can be ignored. Under visible light irradiation, the apparent rate constant k of pure CuTCPP is 0.0012 h−1 , suggesting that the CuTCPP itself hardly has any the visible light activity. All of the CuTCPP/g-C3 N4 composites exhibited higher photocatalytic activity than the single CuTCPP and g-C3 N4 under visible light irradiation. The photocatalyst activity of CuTCPP/g-C3 N4 was enhanced gradually with the increasing proportion of CuTCPP and reaches the optimum activity when the loading amount of CuTCPP is up to 0.75 wt.%. However, as the proportion of CuTCPP further increases, the degradation rate decreases gradually. The apparent k of 0.75% CuTCPP/g-C3 N4 sample is 0.024 h−1 , which was about 2.18 times as high as that of pure g-C3 N4 sample (0.011 h−1 ).
Fig. 2. FT-IR spectra of g-C3 N4 , CuTCPP and CuTCPP/g-C3 N4 photocatalysts.
Fig. 3. Apparent rate constants for the photocatalytic degradation of phenol (a) over g-C3 N4 , CuTCPP and CuTCPP/g-C3 N4 photocatalysts under visible light irradiation ( > 420 nm); (b) over g-C3 N4 and 0.75%CuTCPP/g-C3 N4 photocatalysts under different wavelength light irradiation.
370
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
Fig. 4. The transient photocurrent density responses of pure g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and 0.75%CuTCPP/g-C3 N4 samples electrodes with light on/off cycles under visible light irradiation ( > 420 nm).
The phenol photodegradation intermediates are investigated by chromatograms of phenol before and after photocatalytic degradation for 10 h monitored at 275 nm (Fig. S3). The peak at 5.3 min can be identified as that of phenol. The peak at 3.2, 2.8 and 2.1 min are oxidized intermediates, which might be hydroquinol (HQ), maleic acid (MA) and the lower fatty acid, respectively [34,35]. As the reaction proceeded, the peak of phenol became lower, whereas the peaks of the intermediate species increased gradually. The higher intensity of intermediates by 0.75%CuTCPP/g-C3 N4 than pure gC3 N4 indicates that the degradation ability of 0.75%CuTCPP/g-C3 N4 is remarkably enhanced and the intermediates could be further degraded by ring cleavage and finally subjected to complete degradation to CO2 and H2 O. In order to clearly elucidate the extended photoresponse and enhanced photocatalytic activity of CuTCPP/g-C3 N4 photocatalyst in longer wavelength visible regions, the wavelength dependence of photocatalytic activity for phenol degradation over sensitization photocatalyst investigated in the range of 420–550 nm using various band pass filters ( = 420, 450, 470, 490, 510 and 550 nm). Fig. 3b compared the value of rate constants k of pure g-C3 N4 and 0.75%CuTCPP/g-C3 N4 composite under different ranges of the wavelength. The overall enhanced photocatalytic activity for phenol degradation is observed on 0.75%CuTCPP/g-C3 N4 across the whole of its absorption spectrum. This wavelength-independent enhancement indicates there may be some interactions between CuTCPP and g-C3 N4 that plays an important role in improving the photocatalytic activity. It can be seen that the photocatalytic activity decrease with increasing wavelengths. With the increase the wavelengths from 420 nm to 550 nm, the rate constants of g-C3 N4 decrease from 0.01098 to 0.00045 h−1 , and those of 0.75%CuTCPP/gC3 N4 decrease from 0.02368 to 0.00128 h−1 . This fact is ascribed to the decrease of the available photon energy with the increase of wavelengths, and further confirms that CuTCPP as sensitizer can extend photoresponse of g-C3 N4 and improve utilization efficiency of visible light. The photocurrent responses of 0.75%CuTCPP/g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and pure g-C3 N4 after deposition on ITO electrodes under visible light ( > 420 nm), are shown in Fig. 4. The transient photocurrent responses of a photocatalysis may directly correlate with the recombination efficiency of the photogenerated carriers [36–38]. It is obvious to observe that the photocurrent over CuTCPP/g-C3 N4 is greatly improved, and the optimum 0.75%CuTCPP/g-C3 N4 sample electrode is about 3.5 times as high as that of the pure g-C3 N4 electrode. The enhanced photocurrent over CuTCPP/g-C3 N4 sample implies more efficient separation of the photoinduced electron-hole pairs and longer lifetime of the
Fig. 5. UV–vis diffuse reflection spectra of g-C3 N4 , CuTCPP and CuTCPP/g-C3 N4 photocatalysts.
photogenerated charge carriers than that of pure g-C3 N4 , which is beneficial for its enhanced photocatalytic activity. 3.3. Mechanism of photocatalytic activity enhancement The absorption range of light plays an important role in the photocatalysis, especially for the visible light photodegradation of contaminants. Fig. 5 shows the UV–vis diffuse reflectance spectroscopy (DRS) of CuTCPP/g-C3 N4 composites, as well as pure g-C3 N4 and CuTCPP. The pure g-C3 N4 exhibits a fundamental absorption edge at 450 nm, corresponding to the band gap of 2.7 eV. The pure CuTCPP sample has strong absorption in visible region (400–700 nm) and a strongest absorption intensity peak at 538 nm, which identified as the Q band of porphyrin. As increasing the loading amount of CuTCPP in the CuTCPP/g-C3 N4 composites, the absorption intensities of the composites are enhanced and the absorption edges also show a slight red shift. The CuTCPP/g-C3 N4 composite samples show hybrid absorption features of g-C3 N4 and CuTCPP, which allows for more efficient utilization of the solar spectrum to create photogenerated electrons and holes. Fig. S2 (see the Supporting Information) shows that the Q bands centered at 545 nm attributed to the Q band of CuTCPP adsorbed on the gC3 N4 nanosheet is founded to have a slight red shift of 7 nm as compared to its absorption in methanol at 538 nm, suggesting that – stacking interactions between the individual components in the CuTCPP/g-C3 N4 nanocomposite [39,40]. A significant broadening of the Soret band peak can also be observed in the absorption of CuTCPP solid and CuTCPP absorbed on the g-C3 N4 as compared to the absorption spectra of in methanol. This phenomenon is due to the dye aggregations on the solid substrates [41]. It is well known that phase structure, BET surface area, and separation efficiency of photogenerated charges are the key factor to influent the photocatalytic activity. As can be seen from the XRD spectra, the crystal phase structure of gC3 N4 does not changed after the hybridization of the g-C3 N4 . The BET surface area of g-C3 N4 is 8.2 m2 /g, and those of 0.1%CuTCPP/g-C3 N4 , 0.25%CuTCPP/g-C3 N4 , 0.75%CuTCPP/g-C3 N4 , 0.9%CuTCPP/g-C3 N4 , 1.0%CuTCPP/g-C3 N4 , and 2.0%CuTCPP/g-C3 N4 are 10.1 m2 /g, 9.8 m2 /g, 7.3 m2 /g, 7.2 m2 /g, 7.0 m2 /g, 6.3 m2 /g, respectively. This fact shows that the hybridization with CuTCPP has a little effect in BET surface area of g-C3 N4 . Furthermore, the adsorption experiment shows that adsorption abilities of the g-C3 N4 and CuTCPP/g-C3 N4 nanocomposites toward phenol are almost same. Therefore, the important reason for the enhancement of photocatalytic activity on CuTCPP/g-C3 N4 nanocomposite is due to the effective separation of the photogenerated
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
371
Fig. 6. Photoluminescence spectra of pure g-C3 N4 , pure CuTCPP, 0.25%CuTCPP/gC3 N4 and 0.75%CuTCPP/g-C3 N4 photocatalysts.
electron-hole pairs induced by the energy level match between g-C3 N4 and CuTCPP. Photoluminescence (PL) technique can be also employed to investigate the migration, transfer and recombination processes of photogenerated electron–hole pairs in semiconductors [42]. Fig. 6 presents the photoluminescence spectra for pure g-C3 N4 , 0.75%CuTCPP/g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and pure CuTCPP at an excitation wavelength of 375 nm. Pure g-C3 N4 shows a strong, wide peak in the PL spectrum, while pure CuTCPP has no PL signal response at the excitation wavelength. The PL peak of the CuTCPP/gC3 N4 composites decrease remarkably compared with the pure g-C3 N4 , indicating the recombination of electrons and holes is hindered greatly. Moreover, in comparison with pure g-C3 N4 , the intensity of the PL signal for the CuTCPP/g-C3 N4 composite shows an obvious red-shift, which further indicates that there may be some interactions between CuTCPP and g-C3 N4 . Therefore it can be concluded that CuTCPP have been successfully hybridized with gC3 N4 and the faster separation of photogenerated charges is direct reason for the enhanced photocatalytic activities of CuTCPP/g-C3 N4 composites. To reveal the photocatalytic mechanism further, the interface charge separation efficiency is investigated by the typical electrochemical impedance spectra (presented as Nyquist plots). Fig. 7 shows electrochemical impedance spectra (EIS) Nyquist plots of 0.75%CuTCPP/g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and pure g-C3 N4 electrodes before and after visible light irradiation ( > 420 nm). The arc radius on the EIS Nyquist plot can reflect the reaction rate on the surface of the electrode. This smaller arc radius implies a more effective separation of photogenerated electron–hole pairs
Fig. 7. Nyquist plots for pure g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and 0.75%CuTCPP/gC3 N4 in aqueous solution in the dark and under visible light illumination ( > 420 nm) [Na2 SO4 = 0.1 M].
Fig. 8. The plots of photogenerated carriers trapping in the system of photodegradation of MB by (a) pure g-C3 N4 and (b) 0.75%CuTCPP/g-C3 N4 , under visible light irradiation ( > 420 nm), respectively.
and a higher efficiency of charge immigration across the electrode/electrolyte interface [43]. The arc radii of CuTCPP/g-C3 N4 samples electrode are smaller than that of pure g-C3 N4 electrode, suggesting that CuTCPP/g-C3 N4 composite structure can make the separation and immigration of photogenerated electron-hole pairs more efficient, which is in good accordance with the result of the photocurrent measurement. It is important to detect main oxidative species in the photocatalytic process for elucidating the photocatalytic mechanism. The main oxidative species in photocatalytic process are detected through the trapping experiments of radicals using IPA as hydroxyl radical scavenger [44] and CH2 O2 as holes radical scavenger [45], and purging N2 as ·O2− scavenger[46], respectively. As shown in Fig. 8a, the photocatalytic degradation of phenol with g-C3 N4 hardly has any change by the addition of IPA in the reaction system, while has a remarkably decrease by the addition of CH2 O2 and purging N2 gas, indicating that both ·O2− and photogenerated holes might have the effect on the phenol degradation in the pure gC3 N4 system. For CuTCPP/g-C3 N4 composite, the change tendency is same with the g-C3 N4 . The photocatalytic activity is greatly suppressed by the addition of a scavenger for holes (CH2 O2 ) and by N2 gas in the reaction system (Fig. 8b). While the addition of IPA only causes a slight change in the photodegradation of phenol. This fact indicates that both ·O2− and photogenerated holes are main oxidative species of CuTCPP/g-C3 N4 for phenol degradation under visible light irradiation. However, compared with pure g-C3 N4 , the straight slopes of CuTCPP/g-C3 N4 samples with same addition of CH2 O2 and N2 decline more obviously, indicating that there may exist more holes and ·O2− in CuTCPP/g-C3 N4 system than in pure g-C3 N4 system due to the effective separation of photogenerated electron–hole pairs.
372
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
4. Conclusion In this work, we present a simple method to prepare CuTCPP/g-C3 N4 nanohybrid photocatalyst for the organic compound degradation. Porphyrin can easily hybridize with g-C3 N4 through strong – stacking interactions without much influence on the electronic properties of the g-C3 N4 . Sensitization of g-C3 N4 with CuTCPP affords a highly efficient photocatalyst for phenol degradation under visible light irradiation. When the CuTCPP content in CuTCPP/g-C3 N4 composites achieves 0.75%, the photocatalytic activity is up to the maximum, which is almost 2.2 times as high as that of pure g-C3 N4 under the visible light. Both the holes and ·O2− are main oxidative species of CuTCPP/g-C3 N4 for phenol degradation under visible light irradiation. The enhancement in photocatalytic performance under the visible light irradiation is ascribed to not only the well-matched overlapping band structures, which facilitating the separation and transfer of photogenerated electron–hole pairs, but also the sensitization effect of CuTCPP, which leading to an overall enhanced photocatalytic activity of g-C3 N4 across the whole its absorption spectrum. Acknowledgements This present work is supported by the National Natural Science Foundations of China (Grant No.21106138), the Fundamental Research Funds for the Central Universities (Grant No. 2011YXL062), National High Technology Research and Development Program of China (2012AA062701), Open fund of National Laboratory of Mineral Materials (Grant No. 09A003). Fig. 9. Schematic drawing illustrating the mechanism of charge separation and photocatalytic activity of the CuTCPP/g-C3 N4 photocatalyst under visible light irradiation (a) > 420 nm; (b) > 450 nm.
On the base of the above information, a possible mechanism for the phenol photodegradation using the CuTCPP hybridized g-C3 N4 photocatalyst is proposed as illustrated in Fig. 9. Both the g-C3 N4 and CuTCPP can be excited by visible light ( > 420 nm) and produce photogenerated electrons and holes (Fig. 9 a). Since the conduction band (CB) position of CuTCPP is lower than that of g-C3 N4 , the photoinduced electrons on the CB of CuTCPP can directly transfer to the CB of g-C3 N4 . Consequently, the photoinduced electrons in CB of gC3 N4 are good reducing agents that could capture the adsorbed O2 onto the composite catalyst surface and reduce it to ·O2− . Subsequently, phenol molecules can be degraded by oxidation reaction with ·O2− . Meanwhile, the corresponding valence band (VB) position of g-C3 N4 is lower than that of CuTCPP, the photogenerated holes on the VB of g-C3 N4 can spontaneously immigrate to the VB of CuTCPP. The holes on VB of CuTCPP can directly degrade the organic compounds under visible light irradiation. Therefore, an effective charge separation can be achieved due to the well-matched overlapping band structures and closely contacted interfaces between CuTCPP and g-C3 N4 . This fact reduces the probability of charges recombination, resulting in enhanced photocatalytic activity under visible light irradiation. In addition, when the wavelength is larger than 450 nm (Fig. 9 b), g-C3 N4 cannot be excited. CuTCPP can still be excited due to its strong absorption in visible region (400–700 nm), and produce photogenerated electrons and holes. Consequently the photoinduced electrons on the CB of CuTCPP can transfer to the CB of g-C3 N4 , and react with adsorbed O2 on the surface of g-C3 N4 into ·O2− . The ·O2 can react with H2 O to form ·OH or directly oxidize the organic compound into CO2 and H2 . Therefore, the sensitization g-C3 N4 with CuTCPP can lead to an overall enhanced photocatalytic activity across the whole of its absorption spectrum.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.apcatb. 2014.11.050. References [1] J. Sun, J. Zhang, M. Zhang, M. Antonietti, X. Fu, X. Wang, Nat. Commun. 3 (2012) 1132–1139. [2] F. Goettmann, A. Thomas, M. Antonietti, Angew. Chem. Int. Ed. 46 (2007) 2717–2720. [3] Y. Zhang, A. Thomas, M. Antonietti, X. Wang, J. Am. Chem. Soc. 131 (2009) 50–51. [4] Z.H. Zhang, K. Leinenweber, M. Bauer, L.A.J. Garvie, P.F. McMillan, G.H. Wolf, J. Am. Chem. Soc. 123 (2001) 7788–7796. [5] G. Liu, P. Niu, C.H. Sun, S.C. Smith, Z.G. Chen, G.Q. Lu, H.M. Cheng, J. Am. Chem. Soc. 132 (2010) 11642–11648. [6] X.C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen, M. Antonietti, Nat. Mater. 8 (2009) 76–80. [7] Y. Wang, J.S. Zhang, X.C. Wang, M. Antonietti, H.R. Li, Angew. Chem. Int. Ed. 49 (2010) 3356–3359. [8] Y. Di, X.C. Wang, A. Thomas, M. Antonietti, ChemCatChem 2 (2010) 834–838. [9] G.Z. Liao, S. Chen, X. Quan, H.T. Yu, H.M. Zhao, J. Mater. Chem. 22 (2012) 2721–2726. [10] H.J. Yan, Y. Huang, Chem. Commun. 47 (2011) 4168–4170. [11] Q.J. Xiang, J.G. Yu, M. Jaroniec, J. Phys. Chem. C 115 (2011) 7355–7363. [12] Y.J. Zhang, T. Mori, L. Niu, J.H. Ye, Energ. Environ. Sci. 4 (2011) 4517–4521. [13] S. Min, G. Lu, J. Phys. Chem. C 116 (2012) 19644–19652. [14] T. Kamegawa, S. Matsuura, H. Seto, H. Yamashita, Angew. Chem. 125 (2013) 950–953. [15] M. Latorre-Sánchez, C. Lavorato, M. Puche, V. Fornés, R. Molinari, H. Garcia, Chem. A Eur. J. 18 (2012) 16774–16783. [16] Y. Park, W. Kim, D. Monllor-Satoca, T. Tachikawa, T. Majima, W. Choi, J. Phys. Chem. Lett. 4 (2013) 189–194. [17] R. Abe, K. Hara, K. Sayama, K. Domen, H. Arakaw, J. Photochem. Photobiol. A: Chem. 137 (2000) 63–69. [18] Q.Y. Li, Z.L. Jin, Z.G. Peng, Y.X. Li, S.B. Li, G.X. Lu, J. Phys. Chem. C 111 (2007) 8237–8241. [19] Q.Y. Li, L. Chen, G.X. Lu, J. Phys. Chem. C 111 (2007) 11494–11499. [20] S.X. Min, G.X. Lu, J. Phys. Chem. C 115 (2011) 13938–13945. [21] Y. Wang, J. Hong, W. Zhang, R. Xu, Catal. Sci. Technol. 3 (2013) 1703–1711. [22] H. Yan, Y. Huang, Chem. Commun. 47 (2011) 4168–4170. [23] L. Yu, X. Zhang, C. Zhuang, L. Lin, R. Li, T. Peng, Phys. Chem. Chem. Phys. 16 (2014) 4106–4114.
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373 [24] C. Wang, J. Li, G. Mele, G.-M. Yang, F.-X. Zhang, L. Palmisano, G. Vasapollo, Appl. Catal. B: Environ. 89 (2009) 448–454. [25] D. Chen, D. Yang, J. Geng, J. Zhu, Z. Jiang, Appl. Surf. Sci. 255 (2008) 2879–2884. [26] W.J. Sun, G. Li, M. Yao, F. Jiang, Zhang, Catal. Commun. 16 (2011) 90–93. [27] M. Zhu, Z. Li, B. Xiao, Y. Lu, Y. Du, P. Yang, X. Wang, ACS Appl. Mater. Interfaces 5 (2013) 1732–1740. [28] Y. Chen, Z.-H. Huang, M. Yue, F. Kang, Nanoscale 6 (2014) 978–985. [29] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen, M. Antonietti, Nat. Mater. 8 (2009) 76–80. [30] V. Georgakilas, M. Otyepka, A.B. Bourlinos, V. Chandra, N. Kim, K.C. Kemp, P. Hobza, R. Zboril, K.S. Kim, Chem. Rev. 112 (2012) 6156–6214. [31] Y.C. Zhao, D.L. Yu, H.W. Zhou, Y.J. Tian, O. Yanagisawa, J. Mater. Sci. 40 (2005) 2645–2647. [32] X. Li, J. Zhang, L. Shen, Y. Ma, W. Lei, Q. Cui, G. Zou, Appl. Phys. A 94 (2009) 387–392. [33] L. Liu, D. Ma, H. Zheng, X. Li, M. Cheng, X. Bao, Micropor. Mesopor. Mater. 110 (2008) 216–222. [34] L. Liu, H. Liu, Y.P. Zhao, Y. Wang, Y. Duan, G. Gao, M. Ge, W. Chen, Environ. Sci. Technol. 42 (2008) 2342–2348.
373
[35] Y. Liu, Y. Zhu, J. Xu, X. Bai, R. Zong, Y. Zhu, Appl. Catal. B: Environ. 142–143 (2013) 561–567. [36] H. Liu, S. Cheng, M. Wu, H. Wu, J. Zhang, W. Li, C. Cao, J. Phys. Chem. A 104 (2000) 7016–7020. [37] W.H. Leng, Z. Zhang, J.Q. Zhang, C.N. Cao, J. Phys. Chem. B 109 (2005) 15008–15023. [38] H. Park, W. Choi, J. Phys. Chem. B 107 (2003) 3885–3890. [39] Y. Xu, L. Zhao, H. Bai, W. Hong, C. Li, G. Shi, J. Am. Chem. Soc. 131 (2009) 13490–13497. [40] W. Tu, J. Lei, S. Zhang, H. Ju, Chem. A Eur. J. 16 (2010) 10771–10777. [41] P. Castillero, J.R. Sanchez-Valencia, M. Cano, J.M. Pedrosa, J. Roales, A. Barranco, A.R. Gonzalez-Elipse, ACS Appl. Mater. Interfaces 2 (2010) 712–721. [42] Y.-S. Xu, W.-D. Zhang, ChemCatChem 5 (2013) 2343–2351. [43] S.J. Hong, S. Lee, J.S. Jang, J.S. Lee, Energ. Environ. Sci. 4 (2011) 1781–1787. [44] L.-S. Zhang, K.-H. Wong, H.-Y. Yip, C. Hu, J.C. Yu, C.-Y. Chan, P.-K. Wong, Environ. Sci. Technol. 44 (2010) 1392–1398. [45] T. Tan, D. Beydoun, R. Amal, J. Photochem. Photobiol. A: Chem. 159 (2003) 273–280. [46] C. Pan, Y. Zhu, Environ. Sci. Technol. 44 (2010) 5570–5574.
Contents lists available at ScienceDirect
Applied Catalysis B: Environmental journal homepage: www.elsevier.com/locate/apcatb
Visible light photoactivity enhancement via CuTCPP hybridized g-C3 N4 nanocomposite Daimei Chen a,∗ , Kewei Wang a,b , Wangzhi Hong a,b , Ruilong Zong b , Wenqing Yao b , Yongfa Zhu b,∗ a Beijing Key Laboratory of Materials Utilization of Nonmetallic Minerals and Solid Wastes, National Laboratory of Mineral Materials, School of Materials Sciences and Technology, China University of Geosciences, Beijing 100083, China b Department of Chemistry, Tsinghua University, Beijing, 100084, PR China
a r t i c l e
i n f o
Article history: Received 5 October 2014 Received in revised form 21 November 2014 Accepted 24 November 2014 Available online 29 November 2014 Keyword: Photocatalysis g-C3 N4 Sensitization Porphyrin
a b s t r a c t Cu (II) meso-Tetra (4-carboxyphenyl) porphyrin (CuTCPP) hybridized g-C3 N4 (CuTCPP/g-C3 N4 ) composites have been facilely synthesized through the ethanol dispersion method. CuTCPP molecules as a sensitizer could easily assemble on the surface of g-C3 N4 nanosheets mainly through – stacking interaction. The CuTCPP/g-C3 N4 composites show much higher photocatalytic activity for phenol degradation than pure g-C3 N4 under visible light irradiation. The optimum photocatatlytic activity of the CuTCPP/g-C3 N4 composites with weight ratio of CuTCPP at 0.75% is almost 2.2 times as high as that of pure g-C3 N4 under the visible light. The enhancement of the visible light photocatalytic activity comes from the efficient electrons transfer from photoexcited CuTCPP molecules to g-C3 N4 sheets and more efficiently visible-light harvesting due to the CuTCPP sensitization. Both the holes and ·O2− are main oxidative species of CuTCPP/g-C3 N4 for phenol degradation under visible light irradiation. Finally, the possible charge transfer mechanism of enhanced visible light photocatalytic activity was proposed. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Graphite-like carbon nitride (g-C3 N4 ), a metal-free polymer semiconductor, has attracted plenty of attention due to its potential application in solar energy conversion, photosynthesis, electrocatalysis, bioimaging and biomedical application [1–4]. Since Wang et al. firstly reported that the polymeric g-C3 N4 shows superior H2 or O2 evolution activities via water splitting under visible light irradiation, it is considered as an especially promising photocatalyst for hydrogen gas production and organic pollutants degradation using solar energy due to its simple fabrication, low-cost and high stability under light irradiation in solution with pH = 0–14 [5,6]. However, the bandgap of g-C3 N4 is still as large as 2.7 eV with an absorption edge just at 450 nm, which largely restricts the visible-light utilization efficiency. In addition, the high degree of recombination between photoinduced electrons and holes also blocks its practical application. To overcome these problems, many methods have been proposed to improve the photocatalytic performance of g-C3 N4 , for example, by doping nonmetallic element
∗ Corresponding authors. Tel.: +86 10 82332274; fax: +86 10 82322974. E-mail addresses: [email protected] (D. Chen), [email protected] (Y. Zhu). http://dx.doi.org/10.1016/j.apcatb.2014.11.050 0926-3373/© 2014 Elsevier B.V. All rights reserved.
[7], and coupling g-C3 N4 with metals [8], organic semiconductors [9,10], graphene [11,12], activated by organic dyes [13]. Among the various strategies for visible-light harvesting, dye sensitization is an efficient and widely used route to extend the spectral response region of wide-band-gap semiconductors [14–16]. Many dye sensitized photocatalysts have been developed for H2 production, such as sensitized semiconductors [17,18], sensitized multiwall carbon nanotubes [19] and sensitized reduced graphene oxide [20]. Obviously, dye sensitization of g-C3 N4 can also be an efficient way to enhance the utilization efficiency of visible light. Until now, several dyes such as Erythrosin B (ErB), Eosin Y (EY), poly(3-hexylthiophene) (P3HT) and zinc phthalocyanines (ZnPc) were used as sensitizers of g-C3 N4 to enhance the photocatalytic H2 production activity with considerable visible light utilization efficiency [13,21–23]. Porphyrins as light harvesting materials play an important role in photosynthesis. As they possess good chromophore activities over the solar spectrum and good electron donating properties due to their large p-electron systems, porphyrin compounds have been widely regarded as excellent photosensitizers for photocatalysts. For example, the metalloporphyrin-modified TiO2 system has been reported for the photocatalytic degradation of various kinds of organic pollutants in water under visible light irradiation [24–26]. In generally, metalloporphyrin modified photocatalysts
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
were prepared by covalent interaction between the different functional groups ( COOH, OH, etc.) in the periphery of the porphyrin ring and the surface of photocatalysts. It is widely accepted that the covalent bands can form channels, which can enhance the electron transfer between porphyrins and photocatalysts, and further lead to the delocalization of p* orbits of porphyrins [25]. However, the amount of porphyrins absorbed on the surface of the photocatlaysts is limited because most of porphyrins are difficult to covalently interact with photocatlaysts. Zhu et al. prepared porphyrin uncovalently functionalized reduced graphene oxide (RGO) nanocomposite by immobilizing porphyrin on RGO nanosheets via – stacking interactions. Porphyrin moiety in the nanocomposite acts as a photosensitizer and harvesting irradiation light. This composite shows remarkable enhanced photocatalytic activity under UV–vis light irradiation due to the fast electrons transfer from photoexcited porphyrin molecules to RGO sheets [27,28]. Considering the g-C3 N4 has analogue structure with RGO (planar layers with electron conjugation), it is expected to interact strongly with porphyrins through electrostatic interaction, – stacking, or hydrogen bonding. However, to the best our knowledge, there are no studies about the sensitization of g-C3 N4 with porphyrin, especially the using the dye sensitized g-C3 N4 for the organic containment degradation. Therefore, a systemic investigation of dye sensitized g-C3 N4 and the oxidation mechanic of the dye sensitized g-C3 N4 for the photodegradation needs further studied. In our paper, we reported the preparation, characterization, and photocatalysis of CuTCPP/g-C3 N4 composites. Choosing CuTCPP as a sensitizer to hybridize with g-C3 N4 is not only due to that porphyrin can easily hybridize with g-C3 N4 through strong – stacking interactions, but also due to its nice locally conjugated aromatic structure, as well as its remarkable chemical and thermal stability benefitting for photocatalytic reactions. The optical absorption of g-C3 N4 (<460 nm) can be greatly extended into longer wavelength of visible light region (450–600 nm) by spectral sensitization with porphyrin. This sensitization photocatalyst shows a high activity for phenol degradation under visible light irradiation because the photogenerated electrons of the excited porphyrin can transfer to g-C3 N4 efficiently. The present results show the promising application of porphyrin derivatives in a photocatalytic organic degradation system for more efficiently utilizing solar radiation. 2. Experimental 2.1. Material Dicyandiamide (C2 H4 N4 ) was purchased from Sinopharm Chemical Reagent Corp, P. R. China. Cu (II) meso-Tetra (4carboxyphenyl) porphine (CuTCPP, C48 H28 CuN4 O8 ) was supported by J&K Scientific Ltd. All other reagents used in this research were analytically pure and used without further purification. 2.2. Sample preparation 2.2.1. Synthesis of g-C3 N4 Polycondensation of dicyandiamide was employed to obtain gC3 N4 according to Ref [29]. Typically, 15 g of dicyandiamide was put in a Muffle Furnace and heated to 550 ◦ C for 4 h to complete the reaction. Finally, the yellow products were collected and ground into powder in an agate mortar. 2.2.2. Synthesis of CuTCPP/g-C3 N4 composite photocatalyst The CuTCPP/g-C3 N4 photocatalysts were prepared by liquid ultrasonic route in ethanol. Firstly, 0.5 g g-C3 N4 was added into ethanol and then was placed in an ultrasonic bath for an hour to completely disperse the g-C3 N4 . The CuTCPP ethanol solution (0.1 mg mL−1 ) was added into the above solution and sonicated for
367
an hour. Then the resulting suspension was continually stirred in a fume hood at 60 ◦ C for a night. After completely volatilization of the ethanol, the samples were dried at 100 ◦ C for 8 h. A series of CuTCPP/g-C3 N4 photocatalysts with the different mass ratios of CuTCPP were prepared by this method, which were denoted as X%CuTCPP/g-C3 N4 , where X represents the weight ratios of CuTCPP in total (X% = 0.1%, 0.25%, 0.5%, 0.75%, 0.8%, 0.9%, 1%, 2%). 2.3. Characterization The crystallinity of the as-prepared sample was characterized by X-ray diffraction (XRD) on Bruker D8 Advance X-ray diffractometer at room temperature. Morphologies and structures of the prepared samples were further examined with a HITACHI HT770 transmission electron microscopy (TEM) operated at an accelerating voltage of 100 kV. The UV–vis diffuse reflectance spectra (DRS) of the samples were carried out on a Hitachi U-3010 UV–vis spectrophotometer using BaSO4 as the reference. Fourier transform infrared (FTIR) spectra were carried on a Bruker spectrometer in the frequency range of 4000–600 cm−1 with a resolution of 1 cm−1 . Electrochemical and photoelectrochemical measurements were performed in three-electrode quartz cells with a 0.1 M Na2 SO4 electrolyte solution. Platinum wire was used as the counter electrode, and saturated calomel electrodes (SCE) were used as the reference electrodes, respectively. The as-prepared photocatalyst film electrodes on ITO served as the working electrode. The photocurrents were measured on an electrochemical system (CHI-660B, China). Electrochemical impedance spectra (EIS) were measured at 0.0 V. A sinusoidal ac perturbation of 5 mV was applied to the electrode over the frequency range of 0.05–105 Hz. Photoluminescence spectra (PL) of the samples were obtained at room temperature excited by incident light of 375 nm using a HORIBA Aqualog Fluorescence Spectrometer. 2.4. Photodegradation experiment The photocatalytic activities of the as-prepared samples were evaluated by the decomposition of phenol under visible light irradiation. Visible light source was obtained by a 500 W Xe lamp with different wavelengths (420 nm, 450 nm, 470 nm, 490 nm, 510 nm, 550 nm) cutoff filter and, and the average light intensity was 35 mW/cm2 . 25 mg photocatalyst was added into prepared 50 ml 5 ppm phenol solution. Before the light irradiation, the suspensions were firstly ultrasonic dispersed in dark for 15 min, then magnetically stirred for 120 min to reach the adsorption–desorption equilibrium. At given time intervals, 2.8 ml aliquots were sampled and filtered with a micropore membrane. Synchronously, the filtrates of phenol solutions at different conditions were analyzed by recording variations of the phenol peak area. HPLC was adopted to detect phenol concentration and its degradation products distribution. Methanol and water (volume ratio: 60/40) were mobile phase, the elution time was 7 min, the flow rate was 1 ml/min, determine wavelength was 270 nm and chromatographic column was a Venusil XBP-C18 (3.9 × 200, Agela Technologies Inc.) column. Before the analysis, the samples were filtered through Millipore discs of 0.45 m to protect the chromatographic column. 3. Results and discussion 3.1. Morphology and Structure of CuTCPP hybridized g-C3 N4 photocatalysts Experimentally, the formation of CuTCPP hybridized g-C3 N4 structure was involved involves a two-steps process (Scheme 1). First, g-C3 N4 is exfoliated into sheet structures through an ultrasonic method in ethanol solvent. Secondly, the CuTCPP ethanol solution (0.1 mg mL−1 ) was added into the above solution and
368
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
Scheme 1. Schematic illustration of the preparation of the CuTCPP hybridized g-C3 N4 composites.
bath-sonicated for an hour. Consequently, the CuTCPP molecules will spontaneously coat on the surface of g-C3 N4 nanosheets through the intermolecular interactions, such as electrostatic interaction and – stacking interaction, resulting in a CuTCPP/g-C3 N4 nanohybrid composite. It is well-known that porphyrin molecules, which have a two-dimensional 18--electron aromatic structure, could interact with g-C3 N4 through – stacking. Meanwhile because the functional groups–COOH in the periphery of the CuTCPP ring might interact with the groups NH2 on the edge of the g-C3 N4 , the CuTCPP and g-C3 N4 molecules can also be covalently bonded together by an amide bond. Therefore, both the uncovalently and covalently interactions facilitate the formation of the CuTCPP/g-C3 N4 hybrid nanocomposite. However, owing to the small amount of the groups NH2 on the edge of the g-C3 N4 , the covalent interaction between the CuTCPP and g-C3 N4 molecules might be very little. So, it can be deduced that the hybridization effect between the CuTCPP and g-C3 N4 molecules might mainly come from the noncovalent interaction. The great advantage of the strategy of noncovalent interaction is that it can combine the unique properties of the CuTCPP and g-C3 N4 , while it does not disturb the physical properties of both moieties [30]. Fig. 1 shows transmission electron microscopy (TEM) images of pure g-C3 N4 , CuTCPP and 0.75%CuTCPP/g-C3 N4 . Most of irregular and large plates of g-C3 N4 are observed in Fig. 1a and b. In the image of CuTCPP (Fig. 1 c and d), a lot of small rodlike structure materials can be observed which might be the CuTCPP molecules aggregations. The average size of CuTCPP molecules aggregations is about 50 × 200 nm. For the 0.75%CuTCPP/g-C3 N4 composite (Fig. 1e and f), there are many large plates of g-C3 N4 , while no CuTCPP aggregation can be found in the TEM image. Meanwhile, compared with pure g-C3 N4 , the agglomeration of CuTCPP/g-C3 N4 composite becomes more serious. These observations demonstrate that CuTCPP molecules might be well dispersed and attached on g-C3 N4 plates. To identify the change of phase structures before and after hybridization, the XRD patterns of the as-prepared CuTCPP/g-C3 N4 composites, as well as pure g-C3 N4 and CuTCPP as shown in Figure S1. The g-C3 N4 sample has a characteristic peak at 27.4◦ , which
can be indexed as the (0 0 2) diffraction plane. And it is can be seen that the crystal phase of g-C3 N4 did not change with increasing CuTCPP content. There was no crystalline CuTCPP in the CuTCPP/gC3 N4 photocatalysts. This fact might be that the loading amount of CuTCPP is too low to detect, and it can also be inferred that CuTCPP molecules dispersed uniformly on the surface of g-C3 N4 , which exhibits a short-range order of CuTCPP molecules arrangement. The FT-IR spectra of pure g-C3 N4 , pure CuTCPP and a series of CuTCPP/g-C3 N4 photocatalysts are shown in Fig. 2. In the FT-IR spectrum of g-C3 N4 , the peak at 1637 cm−1 is attributable to the C N stretching vibration modes, while the peaks at 1243 cm−1 , 1320 cm−1 and 1403 cm−1 are due to the aromatic C N stretching [31–33]. The peak at 808 cm−1 is related to the s-triazine ring modes [32]. These peaks are present in the all CuTCPP/g-C3 N4 composites, suggesting that no structural change of g-C3 N4 occurs. It can be seen that there was no characteristic peaks of CuTCPP in the CuTCPP/gC3 N4 photocatalysts at the low CuTCPP loadings (<2%). When the loading amount of CuTCPP is more than 10%, the main characteristic peaks of CuTCPP molecules can be observed clearly, indicating that the CuTCPP molecules might homogenously immobilized on the surface of the g-C3 N4 nanosheets and successfully hybridized with g-C3 N4 . However, the characteristic stretching and stretching band of the amide group formed between the CuTCPP and g-C3 N4 molecules, which are reported to occur at 1640 and 1260, respectively, can not be found in the FT-IR spectra of CuTCPP/g-C3 N4 composite. The likely reason might be that these peaks are too small to be overlapped by the peaks of g-C3 N4 . This further confirms that the hybridization effect between the CuTCPP and g-C3 N4 molecules might mainly come from the noncovalent interaction. 3.2. Photocatalytic activity and photocurrent of CuTCPP hybridized g-C3 N4 composite photocatalysts The photocatalytic performance was mainly evaluated by degradation of phenol, a hazardous pollutant for water as well as a common model to test the photodegradation activity. As shown in Fig. 3a, significant differences in the catalytic behaviors were
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
369
Fig. 1. TEM images of (a, b) pure g-C3 N4 , (c, d) pure CuTCPP, (e, f) 0.75%CuTCPP/g-C3 N4 .
observed, and the photodegradation process was fit to pseudofirst-order kinetics in which the value of rate constant k is equal to the corresponding slope of the fitting line. The blank test confirms that phenol is only slightly degraded in the absence of catalysis, indicating that the photolysis effects can be ignored. Under visible light irradiation, the apparent rate constant k of pure CuTCPP is 0.0012 h−1 , suggesting that the CuTCPP itself hardly has any the visible light activity. All of the CuTCPP/g-C3 N4 composites exhibited higher photocatalytic activity than the single CuTCPP and g-C3 N4 under visible light irradiation. The photocatalyst activity of CuTCPP/g-C3 N4 was enhanced gradually with the increasing proportion of CuTCPP and reaches the optimum activity when the loading amount of CuTCPP is up to 0.75 wt.%. However, as the proportion of CuTCPP further increases, the degradation rate decreases gradually. The apparent k of 0.75% CuTCPP/g-C3 N4 sample is 0.024 h−1 , which was about 2.18 times as high as that of pure g-C3 N4 sample (0.011 h−1 ).
Fig. 2. FT-IR spectra of g-C3 N4 , CuTCPP and CuTCPP/g-C3 N4 photocatalysts.
Fig. 3. Apparent rate constants for the photocatalytic degradation of phenol (a) over g-C3 N4 , CuTCPP and CuTCPP/g-C3 N4 photocatalysts under visible light irradiation ( > 420 nm); (b) over g-C3 N4 and 0.75%CuTCPP/g-C3 N4 photocatalysts under different wavelength light irradiation.
370
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
Fig. 4. The transient photocurrent density responses of pure g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and 0.75%CuTCPP/g-C3 N4 samples electrodes with light on/off cycles under visible light irradiation ( > 420 nm).
The phenol photodegradation intermediates are investigated by chromatograms of phenol before and after photocatalytic degradation for 10 h monitored at 275 nm (Fig. S3). The peak at 5.3 min can be identified as that of phenol. The peak at 3.2, 2.8 and 2.1 min are oxidized intermediates, which might be hydroquinol (HQ), maleic acid (MA) and the lower fatty acid, respectively [34,35]. As the reaction proceeded, the peak of phenol became lower, whereas the peaks of the intermediate species increased gradually. The higher intensity of intermediates by 0.75%CuTCPP/g-C3 N4 than pure gC3 N4 indicates that the degradation ability of 0.75%CuTCPP/g-C3 N4 is remarkably enhanced and the intermediates could be further degraded by ring cleavage and finally subjected to complete degradation to CO2 and H2 O. In order to clearly elucidate the extended photoresponse and enhanced photocatalytic activity of CuTCPP/g-C3 N4 photocatalyst in longer wavelength visible regions, the wavelength dependence of photocatalytic activity for phenol degradation over sensitization photocatalyst investigated in the range of 420–550 nm using various band pass filters ( = 420, 450, 470, 490, 510 and 550 nm). Fig. 3b compared the value of rate constants k of pure g-C3 N4 and 0.75%CuTCPP/g-C3 N4 composite under different ranges of the wavelength. The overall enhanced photocatalytic activity for phenol degradation is observed on 0.75%CuTCPP/g-C3 N4 across the whole of its absorption spectrum. This wavelength-independent enhancement indicates there may be some interactions between CuTCPP and g-C3 N4 that plays an important role in improving the photocatalytic activity. It can be seen that the photocatalytic activity decrease with increasing wavelengths. With the increase the wavelengths from 420 nm to 550 nm, the rate constants of g-C3 N4 decrease from 0.01098 to 0.00045 h−1 , and those of 0.75%CuTCPP/gC3 N4 decrease from 0.02368 to 0.00128 h−1 . This fact is ascribed to the decrease of the available photon energy with the increase of wavelengths, and further confirms that CuTCPP as sensitizer can extend photoresponse of g-C3 N4 and improve utilization efficiency of visible light. The photocurrent responses of 0.75%CuTCPP/g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and pure g-C3 N4 after deposition on ITO electrodes under visible light ( > 420 nm), are shown in Fig. 4. The transient photocurrent responses of a photocatalysis may directly correlate with the recombination efficiency of the photogenerated carriers [36–38]. It is obvious to observe that the photocurrent over CuTCPP/g-C3 N4 is greatly improved, and the optimum 0.75%CuTCPP/g-C3 N4 sample electrode is about 3.5 times as high as that of the pure g-C3 N4 electrode. The enhanced photocurrent over CuTCPP/g-C3 N4 sample implies more efficient separation of the photoinduced electron-hole pairs and longer lifetime of the
Fig. 5. UV–vis diffuse reflection spectra of g-C3 N4 , CuTCPP and CuTCPP/g-C3 N4 photocatalysts.
photogenerated charge carriers than that of pure g-C3 N4 , which is beneficial for its enhanced photocatalytic activity. 3.3. Mechanism of photocatalytic activity enhancement The absorption range of light plays an important role in the photocatalysis, especially for the visible light photodegradation of contaminants. Fig. 5 shows the UV–vis diffuse reflectance spectroscopy (DRS) of CuTCPP/g-C3 N4 composites, as well as pure g-C3 N4 and CuTCPP. The pure g-C3 N4 exhibits a fundamental absorption edge at 450 nm, corresponding to the band gap of 2.7 eV. The pure CuTCPP sample has strong absorption in visible region (400–700 nm) and a strongest absorption intensity peak at 538 nm, which identified as the Q band of porphyrin. As increasing the loading amount of CuTCPP in the CuTCPP/g-C3 N4 composites, the absorption intensities of the composites are enhanced and the absorption edges also show a slight red shift. The CuTCPP/g-C3 N4 composite samples show hybrid absorption features of g-C3 N4 and CuTCPP, which allows for more efficient utilization of the solar spectrum to create photogenerated electrons and holes. Fig. S2 (see the Supporting Information) shows that the Q bands centered at 545 nm attributed to the Q band of CuTCPP adsorbed on the gC3 N4 nanosheet is founded to have a slight red shift of 7 nm as compared to its absorption in methanol at 538 nm, suggesting that – stacking interactions between the individual components in the CuTCPP/g-C3 N4 nanocomposite [39,40]. A significant broadening of the Soret band peak can also be observed in the absorption of CuTCPP solid and CuTCPP absorbed on the g-C3 N4 as compared to the absorption spectra of in methanol. This phenomenon is due to the dye aggregations on the solid substrates [41]. It is well known that phase structure, BET surface area, and separation efficiency of photogenerated charges are the key factor to influent the photocatalytic activity. As can be seen from the XRD spectra, the crystal phase structure of gC3 N4 does not changed after the hybridization of the g-C3 N4 . The BET surface area of g-C3 N4 is 8.2 m2 /g, and those of 0.1%CuTCPP/g-C3 N4 , 0.25%CuTCPP/g-C3 N4 , 0.75%CuTCPP/g-C3 N4 , 0.9%CuTCPP/g-C3 N4 , 1.0%CuTCPP/g-C3 N4 , and 2.0%CuTCPP/g-C3 N4 are 10.1 m2 /g, 9.8 m2 /g, 7.3 m2 /g, 7.2 m2 /g, 7.0 m2 /g, 6.3 m2 /g, respectively. This fact shows that the hybridization with CuTCPP has a little effect in BET surface area of g-C3 N4 . Furthermore, the adsorption experiment shows that adsorption abilities of the g-C3 N4 and CuTCPP/g-C3 N4 nanocomposites toward phenol are almost same. Therefore, the important reason for the enhancement of photocatalytic activity on CuTCPP/g-C3 N4 nanocomposite is due to the effective separation of the photogenerated
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
371
Fig. 6. Photoluminescence spectra of pure g-C3 N4 , pure CuTCPP, 0.25%CuTCPP/gC3 N4 and 0.75%CuTCPP/g-C3 N4 photocatalysts.
electron-hole pairs induced by the energy level match between g-C3 N4 and CuTCPP. Photoluminescence (PL) technique can be also employed to investigate the migration, transfer and recombination processes of photogenerated electron–hole pairs in semiconductors [42]. Fig. 6 presents the photoluminescence spectra for pure g-C3 N4 , 0.75%CuTCPP/g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and pure CuTCPP at an excitation wavelength of 375 nm. Pure g-C3 N4 shows a strong, wide peak in the PL spectrum, while pure CuTCPP has no PL signal response at the excitation wavelength. The PL peak of the CuTCPP/gC3 N4 composites decrease remarkably compared with the pure g-C3 N4 , indicating the recombination of electrons and holes is hindered greatly. Moreover, in comparison with pure g-C3 N4 , the intensity of the PL signal for the CuTCPP/g-C3 N4 composite shows an obvious red-shift, which further indicates that there may be some interactions between CuTCPP and g-C3 N4 . Therefore it can be concluded that CuTCPP have been successfully hybridized with gC3 N4 and the faster separation of photogenerated charges is direct reason for the enhanced photocatalytic activities of CuTCPP/g-C3 N4 composites. To reveal the photocatalytic mechanism further, the interface charge separation efficiency is investigated by the typical electrochemical impedance spectra (presented as Nyquist plots). Fig. 7 shows electrochemical impedance spectra (EIS) Nyquist plots of 0.75%CuTCPP/g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and pure g-C3 N4 electrodes before and after visible light irradiation ( > 420 nm). The arc radius on the EIS Nyquist plot can reflect the reaction rate on the surface of the electrode. This smaller arc radius implies a more effective separation of photogenerated electron–hole pairs
Fig. 7. Nyquist plots for pure g-C3 N4 , 0.25%CuTCPP/g-C3 N4 and 0.75%CuTCPP/gC3 N4 in aqueous solution in the dark and under visible light illumination ( > 420 nm) [Na2 SO4 = 0.1 M].
Fig. 8. The plots of photogenerated carriers trapping in the system of photodegradation of MB by (a) pure g-C3 N4 and (b) 0.75%CuTCPP/g-C3 N4 , under visible light irradiation ( > 420 nm), respectively.
and a higher efficiency of charge immigration across the electrode/electrolyte interface [43]. The arc radii of CuTCPP/g-C3 N4 samples electrode are smaller than that of pure g-C3 N4 electrode, suggesting that CuTCPP/g-C3 N4 composite structure can make the separation and immigration of photogenerated electron-hole pairs more efficient, which is in good accordance with the result of the photocurrent measurement. It is important to detect main oxidative species in the photocatalytic process for elucidating the photocatalytic mechanism. The main oxidative species in photocatalytic process are detected through the trapping experiments of radicals using IPA as hydroxyl radical scavenger [44] and CH2 O2 as holes radical scavenger [45], and purging N2 as ·O2− scavenger[46], respectively. As shown in Fig. 8a, the photocatalytic degradation of phenol with g-C3 N4 hardly has any change by the addition of IPA in the reaction system, while has a remarkably decrease by the addition of CH2 O2 and purging N2 gas, indicating that both ·O2− and photogenerated holes might have the effect on the phenol degradation in the pure gC3 N4 system. For CuTCPP/g-C3 N4 composite, the change tendency is same with the g-C3 N4 . The photocatalytic activity is greatly suppressed by the addition of a scavenger for holes (CH2 O2 ) and by N2 gas in the reaction system (Fig. 8b). While the addition of IPA only causes a slight change in the photodegradation of phenol. This fact indicates that both ·O2− and photogenerated holes are main oxidative species of CuTCPP/g-C3 N4 for phenol degradation under visible light irradiation. However, compared with pure g-C3 N4 , the straight slopes of CuTCPP/g-C3 N4 samples with same addition of CH2 O2 and N2 decline more obviously, indicating that there may exist more holes and ·O2− in CuTCPP/g-C3 N4 system than in pure g-C3 N4 system due to the effective separation of photogenerated electron–hole pairs.
372
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373
4. Conclusion In this work, we present a simple method to prepare CuTCPP/g-C3 N4 nanohybrid photocatalyst for the organic compound degradation. Porphyrin can easily hybridize with g-C3 N4 through strong – stacking interactions without much influence on the electronic properties of the g-C3 N4 . Sensitization of g-C3 N4 with CuTCPP affords a highly efficient photocatalyst for phenol degradation under visible light irradiation. When the CuTCPP content in CuTCPP/g-C3 N4 composites achieves 0.75%, the photocatalytic activity is up to the maximum, which is almost 2.2 times as high as that of pure g-C3 N4 under the visible light. Both the holes and ·O2− are main oxidative species of CuTCPP/g-C3 N4 for phenol degradation under visible light irradiation. The enhancement in photocatalytic performance under the visible light irradiation is ascribed to not only the well-matched overlapping band structures, which facilitating the separation and transfer of photogenerated electron–hole pairs, but also the sensitization effect of CuTCPP, which leading to an overall enhanced photocatalytic activity of g-C3 N4 across the whole its absorption spectrum. Acknowledgements This present work is supported by the National Natural Science Foundations of China (Grant No.21106138), the Fundamental Research Funds for the Central Universities (Grant No. 2011YXL062), National High Technology Research and Development Program of China (2012AA062701), Open fund of National Laboratory of Mineral Materials (Grant No. 09A003). Fig. 9. Schematic drawing illustrating the mechanism of charge separation and photocatalytic activity of the CuTCPP/g-C3 N4 photocatalyst under visible light irradiation (a) > 420 nm; (b) > 450 nm.
On the base of the above information, a possible mechanism for the phenol photodegradation using the CuTCPP hybridized g-C3 N4 photocatalyst is proposed as illustrated in Fig. 9. Both the g-C3 N4 and CuTCPP can be excited by visible light ( > 420 nm) and produce photogenerated electrons and holes (Fig. 9 a). Since the conduction band (CB) position of CuTCPP is lower than that of g-C3 N4 , the photoinduced electrons on the CB of CuTCPP can directly transfer to the CB of g-C3 N4 . Consequently, the photoinduced electrons in CB of gC3 N4 are good reducing agents that could capture the adsorbed O2 onto the composite catalyst surface and reduce it to ·O2− . Subsequently, phenol molecules can be degraded by oxidation reaction with ·O2− . Meanwhile, the corresponding valence band (VB) position of g-C3 N4 is lower than that of CuTCPP, the photogenerated holes on the VB of g-C3 N4 can spontaneously immigrate to the VB of CuTCPP. The holes on VB of CuTCPP can directly degrade the organic compounds under visible light irradiation. Therefore, an effective charge separation can be achieved due to the well-matched overlapping band structures and closely contacted interfaces between CuTCPP and g-C3 N4 . This fact reduces the probability of charges recombination, resulting in enhanced photocatalytic activity under visible light irradiation. In addition, when the wavelength is larger than 450 nm (Fig. 9 b), g-C3 N4 cannot be excited. CuTCPP can still be excited due to its strong absorption in visible region (400–700 nm), and produce photogenerated electrons and holes. Consequently the photoinduced electrons on the CB of CuTCPP can transfer to the CB of g-C3 N4 , and react with adsorbed O2 on the surface of g-C3 N4 into ·O2− . The ·O2 can react with H2 O to form ·OH or directly oxidize the organic compound into CO2 and H2 . Therefore, the sensitization g-C3 N4 with CuTCPP can lead to an overall enhanced photocatalytic activity across the whole of its absorption spectrum.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.apcatb. 2014.11.050. References [1] J. Sun, J. Zhang, M. Zhang, M. Antonietti, X. Fu, X. Wang, Nat. Commun. 3 (2012) 1132–1139. [2] F. Goettmann, A. Thomas, M. Antonietti, Angew. Chem. Int. Ed. 46 (2007) 2717–2720. [3] Y. Zhang, A. Thomas, M. Antonietti, X. Wang, J. Am. Chem. Soc. 131 (2009) 50–51. [4] Z.H. Zhang, K. Leinenweber, M. Bauer, L.A.J. Garvie, P.F. McMillan, G.H. Wolf, J. Am. Chem. Soc. 123 (2001) 7788–7796. [5] G. Liu, P. Niu, C.H. Sun, S.C. Smith, Z.G. Chen, G.Q. Lu, H.M. Cheng, J. Am. Chem. Soc. 132 (2010) 11642–11648. [6] X.C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen, M. Antonietti, Nat. Mater. 8 (2009) 76–80. [7] Y. Wang, J.S. Zhang, X.C. Wang, M. Antonietti, H.R. Li, Angew. Chem. Int. Ed. 49 (2010) 3356–3359. [8] Y. Di, X.C. Wang, A. Thomas, M. Antonietti, ChemCatChem 2 (2010) 834–838. [9] G.Z. Liao, S. Chen, X. Quan, H.T. Yu, H.M. Zhao, J. Mater. Chem. 22 (2012) 2721–2726. [10] H.J. Yan, Y. Huang, Chem. Commun. 47 (2011) 4168–4170. [11] Q.J. Xiang, J.G. Yu, M. Jaroniec, J. Phys. Chem. C 115 (2011) 7355–7363. [12] Y.J. Zhang, T. Mori, L. Niu, J.H. Ye, Energ. Environ. Sci. 4 (2011) 4517–4521. [13] S. Min, G. Lu, J. Phys. Chem. C 116 (2012) 19644–19652. [14] T. Kamegawa, S. Matsuura, H. Seto, H. Yamashita, Angew. Chem. 125 (2013) 950–953. [15] M. Latorre-Sánchez, C. Lavorato, M. Puche, V. Fornés, R. Molinari, H. Garcia, Chem. A Eur. J. 18 (2012) 16774–16783. [16] Y. Park, W. Kim, D. Monllor-Satoca, T. Tachikawa, T. Majima, W. Choi, J. Phys. Chem. Lett. 4 (2013) 189–194. [17] R. Abe, K. Hara, K. Sayama, K. Domen, H. Arakaw, J. Photochem. Photobiol. A: Chem. 137 (2000) 63–69. [18] Q.Y. Li, Z.L. Jin, Z.G. Peng, Y.X. Li, S.B. Li, G.X. Lu, J. Phys. Chem. C 111 (2007) 8237–8241. [19] Q.Y. Li, L. Chen, G.X. Lu, J. Phys. Chem. C 111 (2007) 11494–11499. [20] S.X. Min, G.X. Lu, J. Phys. Chem. C 115 (2011) 13938–13945. [21] Y. Wang, J. Hong, W. Zhang, R. Xu, Catal. Sci. Technol. 3 (2013) 1703–1711. [22] H. Yan, Y. Huang, Chem. Commun. 47 (2011) 4168–4170. [23] L. Yu, X. Zhang, C. Zhuang, L. Lin, R. Li, T. Peng, Phys. Chem. Chem. Phys. 16 (2014) 4106–4114.
D. Chen et al. / Applied Catalysis B: Environmental 166–167 (2015) 366–373 [24] C. Wang, J. Li, G. Mele, G.-M. Yang, F.-X. Zhang, L. Palmisano, G. Vasapollo, Appl. Catal. B: Environ. 89 (2009) 448–454. [25] D. Chen, D. Yang, J. Geng, J. Zhu, Z. Jiang, Appl. Surf. Sci. 255 (2008) 2879–2884. [26] W.J. Sun, G. Li, M. Yao, F. Jiang, Zhang, Catal. Commun. 16 (2011) 90–93. [27] M. Zhu, Z. Li, B. Xiao, Y. Lu, Y. Du, P. Yang, X. Wang, ACS Appl. Mater. Interfaces 5 (2013) 1732–1740. [28] Y. Chen, Z.-H. Huang, M. Yue, F. Kang, Nanoscale 6 (2014) 978–985. [29] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen, M. Antonietti, Nat. Mater. 8 (2009) 76–80. [30] V. Georgakilas, M. Otyepka, A.B. Bourlinos, V. Chandra, N. Kim, K.C. Kemp, P. Hobza, R. Zboril, K.S. Kim, Chem. Rev. 112 (2012) 6156–6214. [31] Y.C. Zhao, D.L. Yu, H.W. Zhou, Y.J. Tian, O. Yanagisawa, J. Mater. Sci. 40 (2005) 2645–2647. [32] X. Li, J. Zhang, L. Shen, Y. Ma, W. Lei, Q. Cui, G. Zou, Appl. Phys. A 94 (2009) 387–392. [33] L. Liu, D. Ma, H. Zheng, X. Li, M. Cheng, X. Bao, Micropor. Mesopor. Mater. 110 (2008) 216–222. [34] L. Liu, H. Liu, Y.P. Zhao, Y. Wang, Y. Duan, G. Gao, M. Ge, W. Chen, Environ. Sci. Technol. 42 (2008) 2342–2348.
373
[35] Y. Liu, Y. Zhu, J. Xu, X. Bai, R. Zong, Y. Zhu, Appl. Catal. B: Environ. 142–143 (2013) 561–567. [36] H. Liu, S. Cheng, M. Wu, H. Wu, J. Zhang, W. Li, C. Cao, J. Phys. Chem. A 104 (2000) 7016–7020. [37] W.H. Leng, Z. Zhang, J.Q. Zhang, C.N. Cao, J. Phys. Chem. B 109 (2005) 15008–15023. [38] H. Park, W. Choi, J. Phys. Chem. B 107 (2003) 3885–3890. [39] Y. Xu, L. Zhao, H. Bai, W. Hong, C. Li, G. Shi, J. Am. Chem. Soc. 131 (2009) 13490–13497. [40] W. Tu, J. Lei, S. Zhang, H. Ju, Chem. A Eur. J. 16 (2010) 10771–10777. [41] P. Castillero, J.R. Sanchez-Valencia, M. Cano, J.M. Pedrosa, J. Roales, A. Barranco, A.R. Gonzalez-Elipse, ACS Appl. Mater. Interfaces 2 (2010) 712–721. [42] Y.-S. Xu, W.-D. Zhang, ChemCatChem 5 (2013) 2343–2351. [43] S.J. Hong, S. Lee, J.S. Jang, J.S. Lee, Energ. Environ. Sci. 4 (2011) 1781–1787. [44] L.-S. Zhang, K.-H. Wong, H.-Y. Yip, C. Hu, J.C. Yu, C.-Y. Chan, P.-K. Wong, Environ. Sci. Technol. 44 (2010) 1392–1398. [45] T. Tan, D. Beydoun, R. Amal, J. Photochem. Photobiol. A: Chem. 159 (2003) 273–280. [46] C. Pan, Y. Zhu, Environ. Sci. Technol. 44 (2010) 5570–5574.